I have two text files f1.txt and f2.txt, each one has more than two millians of lines of text. I want to read each line in file f1.txt and grep for that in file f2.txt and if not matching then put the unmatched line in to another file f3.txt. I tried the following way but getting errors, I appriciate if any one can put some input on this. Thanks Reuben while read line do if [ 'grep ${line} f2.txt' -lt 0 ] ; then echo ${line} fi done < f1.txt > f3.txt

Hi: You're main problem was your choice of command substitution marks. If using Bourne it's ` not ' I'm using ksh: #!/bin/ksh while read line do if [[ $(grep -c ${line} f2.txt) == 0 ]] ; then echo ${line} fi done < f1.txt > f3.txt # end script You probably know, this is going to take some time with millions of records. Regards, Nails

Nails Thank you so much, it is working fine. Reuben